Until somebody has an actual system and does a tear down it's all speculation at this point. Of course nobody outside core hobby gamers cares in the first place. But it wil be nice to have some actual confirmed specs.
Yup I agree with you , but Im kinda 95% sure it wont deviate a lot from the rumored specs here.
Why!? I personally LOOVED the game, I couldn't put the game down until I 100% everything.
Looking at the Cyberpunk footage it's pretty clear this is impressive for a handheld costing £395.
I owned one of the first Steam Decks and ended up selling it because the battery life killed the fun for me. Even playing something like AC2 it barely managed 90 minutes, on the Switch it was 5 hours. The fan could get noisy too.
Possibly. I like the handheld feature and might use it more than the original due to the 1080p screen. But docked is the way to go for me. My LG C series OLED is waiting. I am hoping for a 4k or at least HD home scree this time round. Although the LG does a pretty decent job of upscaling the 720p screen. The fan in the dock gets my interest a lot more than anything else. We must be looking at much better clock speeds this time round. And perhaps Nvidia have delivered the goods with their AI DLSS.
The steamdeck is hardly a Switch competitor. Even they state they are not. For the vast majority of people that will buy a Switch 2 this will be their very first experiance of this game. They are simply not core hobby gamers that spend lots of time playing and discussing videogames. The just buy games play them and enjoy.
I'm willing to make a prediction here - it'll look noticeably better on Switch 2 than the Steam Deck.
It already does after a very short development cycle.
The copium is strong with this one.
Until SD 2 drops holiday season haha.
Im talking pure technical specs , not sales
When doecked?! Highly probable. Handheld its gonna be similar to Steam Deck which is my point of argument.
Specs are fun to discuss. I was merely suggesting that outside of the core it's not an important factor.
And I'm saying handheld. As impressive as the Steam Deck's APU is, RDNA 2 is limited.
The Switch 2 delivers excellent performance, and that's what truly matters. Technical comparisons are ultimately irrelevant, especially when comparing a home console from the previous generation to Nintendo's latest hybrid/portable system. Furthermore, there are areas where the Switch 2 outperforms, no one has mentioned any inferiority in absolute terms.
Ampere has substantially better colour compression than GCN, so the figures are not directly comparable. For example the 3060 has 360 GB/s for 13.7 TF, which is 26 GB/s per TF. Then a 1.71 TF Ampere part would need ~44 GB/s by this metric, leaving 24 GB/s for the CPU.
I think we will have similar situation we had with series s vs xbox one x, older console had terrible cpu but massive gpu, while newer one has smaller/weaker gpu but much more capable cpu and much more modern feature set(and an ssd).
And we have someone here saying it won't even reach Xbox 1 level in portable mode, while the One can't even run the game properly
People hate on jaguar, but a few traditionally 30 fps franchises went 60 fps on ps4/xone, Battlefield, Battlefront, Gears of war, halo, Uncharted multiplayer and othersI think we will have similar situation we had with series s vs xbox one x, older console had terrible cpu but massive gpu, while newer one has smaller/weaker gpu but much more capable cpu and much more modern feature set(and an ssd).
Same thing gonna be with ps4pr0 vs switch2 docked or ps4 base vs switch2 in handheld mode.
We forgot how terrible really were those jaguar cores we had in xbox one/s/x and ps4 base/pr0.
Remember the ubisoft slide when it compared ps3 cpu power to ps4 cpu power?
It was simply best console form factor cpu amd had to offer at that time, things couldnt be helped, luckily for the industry amd stopped slacking cpu wise, now we need to worry about gpu's coz modern console cpu's are both very performant and relatively affordable(compared to the gpu's
).
The fuck are you talking about WRT the weather? Decima most definitely has a dynamic weather system. Volumetric cloud simulation even. And the type of weather that can occur depends on the location/elevation (you won't get sandstorms in the arctic regions, or snow in the desert, etc.)
With the 100GB/s it'll always be worse (as in - not really close) than Series S.
It did have better multithreading, but boy did Halo look bad on the xbox one
It didn't look that bad, decent for a 60 fps game
Exactly. Most people think that PS4 is confined to 1.6 GHz Jaguars in physics calculations out of ignorance of its architecture. The APU was heavily tailored towards GPGPU via specific hardware customizations (Onion+ bus, Volatile bit, 8 ACEs), In a way it was designed to assist the CPU in this area, ideally asynchronously.
I wholeheartedly agree with your assessment of Switch 2 being at XB1 level in raw throughput as a handheld. Just to note, XB1's 200 GB/s figure for the 32 MB ESRAM is mostly meaningless damage control/marketing trick which combines reads and writes speeds (the official figure at reveal was 102 GB/s, before slight upclock). They stated that they reached upto ~140 GB/s in a specific tailored application afterwards, but in actual games i very much doubt it. I think PS4's 176 GB/s pool offered about twice as bandwidth in practice at least even with XB1's ESRAM help (the vast majority of RAM being 68 GB/s).
It would've been interesting to see how things played out if Sony was unable to increase the amount of RAM just before launch, 4 gigs of fast ram vs 8 gigs of slow ram. As it was the difference was in line with the 40% better GPU metric, coming down to 900p vs 1080p games at the same quality.
True but the novelty of playing something like Witcher 3 on the go is also a draw. Personally for me I'd only be interested in those titles if they hit 60fps, otherwise I might as well wait until I get home and play on pc.
It should be, its 8 years later.
Doesn't this assessment assume that RTX 3060 has 'ideal' bandwidth to compute ratio as a starting point, is this realistic?
In general difference between XB1 vs PS4 came down to either 900P vs 1080P (44% higher resolution on PS4) or as often seen 720P vs 900P (56% higher on PS4 in this case) with sometimes higher performance on PS4 in GPU/bandwidth intensive scenarios on top. There also cases with some higher settings on PS4. The 40% figure for GPU is for compute only which doesn't take things like twice the ROPs (pixel fill rate) and ASYNC into account. To be fair i don't think third party titles pushed GPGPU side of PS4 that hard.
A similar ratio is maintained across the entire 3000 series line, so it is not specific to the 3060. And the claim is not that this is "ideal", but merely that it should be sufficient to see performance comparable to other Ampere parts.
Indeed. MGS5 was running at locked 60fps on PS4 at 1080p while it was ~25fps on X360 (and ~20fps on PS3 at sub 720p resolution). That legendary Ubisoft benchmark didn't take into account the fact that the Cell was here used like a GPGPU. But in traditionnal CPU tasks ? It was trash (worse than X360 CPU) and Jaguar was like 2.5 times faster.
